Semantic Annotation of Tabular Data for Machine-to-Machine Interoperability via Neuro-Symbolic Anchoring

Shervin Mehryar; Remzi Celebi

Semantic Annotation of Tabular Data for Machine-to-Machine Interoperability via Neuro-Symbolic Anchoring

^*Corresponding author for this work

Research output: Contribution to journal › Conference article in journal › Academic › peer-review

Abstract

In this paper we investigate automated annotation of tabular data using semantic technologies in combination with neural network embedding. Specifically, we propose an anchoring model in which property and cell types from the data embedding space are aligned with ontology relation and entity types. We show that by combining the power of symbolic reasoning, neural embeddings, and loss function design, a significant performance improvement as high as 86% for column property, 82% for column type, and 87% for column qualifier annotations can be achieved based on DBpedia and Wikidata table extractions.

Original language	English
Pages (from-to)	61-71
Number of pages	11
Journal	CEUR Workshop Proceedings
Volume	3557
Publication status	Published - 1 Jan 2023
Event	2023 Semantic Web Challenge on Tabular Data to Knowledge Graph Matching, SemTab 2023 - Athens, Greece Duration: 6 Nov 2023 → 10 Nov 2023 https://sem-tab-challenge.github.io/2023/

Keywords

Interoperability
Neuro-symbolic AI
Semantic Annotation
Tabular data

Access to Document

https://ceur-ws.org/Vol-3557/paper5.pdfLicence: CC BY

Cite this

@article{428b715b423245d2a2d0246443ec90a8,

title = "Semantic Annotation of Tabular Data for Machine-to-Machine Interoperability via Neuro-Symbolic Anchoring",

abstract = "In this paper we investigate automated annotation of tabular data using semantic technologies in combination with neural network embedding. Specifically, we propose an anchoring model in which property and cell types from the data embedding space are aligned with ontology relation and entity types. We show that by combining the power of symbolic reasoning, neural embeddings, and loss function design, a significant performance improvement as high as 86% for column property, 82% for column type, and 87% for column qualifier annotations can be achieved based on DBpedia and Wikidata table extractions.",

keywords = "Interoperability, Neuro-symbolic AI, Semantic Annotation, Tabular data",

author = "Shervin Mehryar and Remzi Celebi",

note = "Funding Information: The authors thank the challenge organizers for their timely and helpful response to inquiries, and the reviewers for their valuable comments. This work is supported by the Horizon Europe Framework Program ?Grant agreement No: 101057062), project AIDAVA ? AI powered Data Curation & Publishing Virtual Assistant). Publisher Copyright: {\textcopyright} 2023 CEUR-WS. All rights reserved.; 2023 Semantic Web Challenge on Tabular Data to Knowledge Graph Matching, SemTab 2023, SemTab 2023 ; Conference date: 06-11-2023 Through 10-11-2023",

year = "2023",

month = jan,

day = "1",

language = "English",

volume = "3557",

pages = "61--71",

journal = "CEUR Workshop Proceedings",

issn = "1613-0073",

publisher = "Rheinisch-Westfaelische Technische Hochschule Aachen * Lehrstuhl Informatik V",

url = "https://sem-tab-challenge.github.io/2023/",

}

TY - JOUR

T1 - Semantic Annotation of Tabular Data for Machine-to-Machine Interoperability via Neuro-Symbolic Anchoring

AU - Mehryar, Shervin

AU - Celebi, Remzi

N1 - Funding Information: The authors thank the challenge organizers for their timely and helpful response to inquiries, and the reviewers for their valuable comments. This work is supported by the Horizon Europe Framework Program ?Grant agreement No: 101057062), project AIDAVA ? AI powered Data Curation & Publishing Virtual Assistant). Publisher Copyright: © 2023 CEUR-WS. All rights reserved.

PY - 2023/1/1

Y1 - 2023/1/1

N2 - In this paper we investigate automated annotation of tabular data using semantic technologies in combination with neural network embedding. Specifically, we propose an anchoring model in which property and cell types from the data embedding space are aligned with ontology relation and entity types. We show that by combining the power of symbolic reasoning, neural embeddings, and loss function design, a significant performance improvement as high as 86% for column property, 82% for column type, and 87% for column qualifier annotations can be achieved based on DBpedia and Wikidata table extractions.

AB - In this paper we investigate automated annotation of tabular data using semantic technologies in combination with neural network embedding. Specifically, we propose an anchoring model in which property and cell types from the data embedding space are aligned with ontology relation and entity types. We show that by combining the power of symbolic reasoning, neural embeddings, and loss function design, a significant performance improvement as high as 86% for column property, 82% for column type, and 87% for column qualifier annotations can be achieved based on DBpedia and Wikidata table extractions.

KW - Interoperability

KW - Neuro-symbolic AI

KW - Semantic Annotation

KW - Tabular data

M3 - Conference article in journal

SN - 1613-0073

VL - 3557

SP - 61

EP - 71

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

T2 - 2023 Semantic Web Challenge on Tabular Data to Knowledge Graph Matching, SemTab 2023

Y2 - 6 November 2023 through 10 November 2023

ER -